|
EVALUATING THE WORK OF PARENT
AIDES
For years parent aide programs have been
providing services to families at risk or involved in child
abuse and neglect. One of the earliest studies conducted in
the '70s cited parent aides as one of the most promising programs.
Studies throughout the years have proved this anecdotally
or through outcome data but none have had a thorough comparison
with a control group. Research has been done in a field close
to parent aides- that of early intervention with families
such as nurse home visitation and Healthy Families. The results
have been generally good in terms of short-term and long-term
effects in helping families and their children. While parent
aides can work with similar families the clear difference
in clientele is that parent aides work with families at risk
or already involved (substantiated) with child abuse - in
other words- families in crisis and with multiple dysfunctions.
Pilot Mountain, NC
I have had a long history with parent aide
programs. In 1981 I helped found a large effort in North Carolina
out of Winston-Salem. Since then I have helped found 16 other
parent aide programs in North Carolina and Virginia as well
as helped found the National Parent Aide Association now known
as the National Parent Aide Network (NPAN). The model that
we use is that of the parent aide programs of the National
Exchange Club, which have 79 programs nationwide www.preventchildabuse.com/parent.htm.
These were founded from SCAN of Arkansas, which provided parent
aide (known there as Lay Therapy) services to 15 counties
in Arkansas since 1972. Sharon Pallone was the founder and
I see her as the originator of one of the two strains of parent
aide work in the US the other being Drs. Ruth and Henry Kempe
in Denver, Colorado. I have been lucky enough to have discussions
with Ruth Kempe, to be with Sharon Pallone at the 10th anniversary
of that program in Little Rock and to have discussed the earliest
research with Ann Cohn, the former national leader of Prevent
Child Abuse who was one of the authors. My own outcome research
of 200 families over 15 years has pushed me to want to know
more about the effect of parent aides and why, I believe,
they are effective.
Winston-Salem, NC
I have combined efforts with Neil Guterman,
PhD. at Columbia University School of Social Work in N.Y.C.,
who is well known for his research with early home visitation
to take a scientific look at parent aides. His book Stopping
Child Maltreatment Before it Starts book for sale at Amazon
has helped all of us as direct service providers understand
what works and what doesn't work with home visitation. His
scientific objectivity and strong research skills will help
those of us providing parent aide services across the country
do a much better job of serving families. We will examine
three items: 1) a controlled study of parent aides and their
effect in providing a) safety for children b) better parenting
c) better discipline and d) social support, 2) the reasons
that parent aides are effective and 3) the cost benefits of
parent aides.
SCAN Parent Aides
I invite you to follow this research which
will be chronicled at this site to learn side by side as we
conduct this evaluation. The tools and processes used in this
research are being shared and scrutinized by my peers in the
field and we would welcome your comments and input. The results
could make a tremendous difference in how we help children
born into one of the most devastating social situations -
CHILD ABUSE.
George M. Bryan Jr.
Executive Director
Exchange SCAN
Winston-Salem, NC
2004
OCAN RESEARCH GRANT
Table of Contents:
1.Objectives
And Need For Assistance
2. Description Of Exchange SCAN - Other
program being evaluated
3. Review of the Research on Parent Aide
Program
4. Conceptual Framework for the Evaluation
5. Contextual factors affecting implementation;
opportunities and barriers
6. Results and Benefits from the Evaluation
7. Approach
8. Evaluation Methodology
9. Variables and Their Measurement
10. Process Variables and Their Measurement
11. Cost Variables and Their Measurement
12. Data Analysis Plan
13. Organizational Profile
14. Roles and Responsibilities
15. Dissemination
16. Bibliography
17. Service Log
CRITERION
1: OBJECTIVES AND NEED FOR ASSISTANCE
1. Goals and objectives of proposed
evaluation project.
Goals
Objectives
1) Evaluate the efficacy of parent aides to prevent the recurrence
of child abuse and neglect.
a) Hire & train data collection staff.
b) Identify eligible families.
c) Prepare CASI protocol.
d) Submit IRB protocol.
e) Recruit eligible families into study.
f) Randomly place families into treatment
or control group.
g) Conduct baseline, 6-month & 12-month
interviews.
h) Develop two data files at 18- and 36-months.
i) Conduct interim data analyses during
months 18 and 19.
j) Conduct final data analyses.
k) Prepare final summary report; manuscripts.
l) Disseminate findings.
2) Determine service delivery costs.
a) Hire & train data collection staff.
b) Gather information from center administrative
records.
c) Calculate total costs of service per
case.
d) Gather information from control and treatment
families on benefits associated with service.
e) Gather information from public records
on control and treatment family expenses; service use; encounters
with authorities, use of public services.
f) Consult with Professor Irwin Garfinkel
(Columbia Univ.).
g) Carry out cost-benefit analyses.
h) Prepare final summary report; manuscripts.
i) Disseminate findings.
3) Identify predictors of success in parent aide services.
a) Hire & train data collection staff.
b) Gather information from center client
files (e.g. demographic data, intervention process).
c) Analyze data to assess degree to which
family enrolled in prospective evaluation are similar.
d) Prepare final summary report; manuscripts.
e) Disseminate findings.
2.
Description of the prevention program to be evaluated and
reasons evaluation merited.
The National Exchange Club Foundation (NECF)
will conduct a prospective evaluation of the parent aide program
located in Winston-Salem, NC. Parent aides are defined by
the National Parent Aide Network (NPAN) as “trained,
professionally supervised individuals, volunteer or paid,
who assist parents under stress and those whose children are
at-risk of abuse or neglect” (NPAN brochure, 2000).
NPAN links parent aide programs in the U.S. Over 600 parent
aide programs (Bryan, 1992) serve an important role in stopping
child abuse yet there are no long-term evaluations of this
model. Recent evaluations of home visiting programs have left
out the role of parent aide programs, e.g. The Packard Foundation’s
report on “Home Visiting: Recent Program Evaluations”
(Packard Foundation, 1999).
The NECF coordinates the largest collection
of parent aide programs in the U.S. with 79 centers serving
over 100 sites. Consistency in programming and adherence to
training modules, supervision, documentation, and closure
procedures over the past 20 years provides stable programming
for evaluation. Findings from this study will have implication
within the field of child abuse prevention by advancing evidenced-based
practice in the field and expanding the current knowledge
base (please see Criterion 1.7).
History. Parent aide programs developed
in the early 1970s through the efforts of Ruth and Henry Kempe
in Denver, Colorado (Helfer and Kempe, 1972, pg. 177-184)
and the work of Sharon Pallone in Little Rock, Arkansas. Berkeley
Planning Associates studied the Little Rock program in one
of the earliest home visiting evaluation reports (Berkeley
Planning Associates and Cohn, 1974-1977). In 1979, the National
Exchange Club (a national service organization) created NECF
as a 501(c) (3) for the purpose of guiding local Exchange
Club groups in establishing parent aide programs. The
first child abuse prevention centers were established in Florida,
Texas, Mississippi, North Carolina, and Maryland in 1981.
NPAN came under the umbrella of the NECF in 1994.
Significant features and components; goals
and objectives of prevention program. To assure consistency
and quality program implementation, NECF provides centers
within its network: parent aide training materials, intensive
orientation for center directors, training symposiums, technical
support, and National Standards of Operation and Practice
(SOP). The SOP focuses on best practice standards in six areas:
Mission; Financial Resources and Management; Human Resources;
Networking/Partnerships; Organizational Resources; [Parent
Aide] Program Practices. Centers are expected to comply with
these standards in order to maintain certification. Individually,
centers develop case policy and procedures and implement standard
parent aide training to assure consistency in service delivery.
Families receiving parent aide services
are referred by a variety of community sources. Referral criteria
are outlined below. Sites maintain policies that outline methods
of accepting and prioritizing referrals. Parent aide services
are provided as follows: 1) Families receive home visits by
professional staff and complete an Initial Needs Assessment
(INA) describing family dynamics and strengths, patterns of
coping and abuse histories, and immediate needs. 2) An initial
family treatment plan is developed and revised every three
months by the family, parent aide, and supervisor. The plan
is family-centered and details and goals center on a) child
safety, b) problem solving skills, c) parenting skills, and
d) social support. 3) Families are matched with a professionally
supervised parent aide who provides weekly documented home
visits. 4) Closure is considered when a level of success occurs
in all four-goal areas and the families’ ability to
handle problems is demonstrated over an additional three months.
Average length of parent aide involvement is 13 months. Parent
aides are screened through a five-part interview and background
check, receive initial 12-hour training session, and receive
weekly supervision. Professional staff is available 24 hours
per day to the parent aide or family to respond to crises.
Characteristics of target population. Families
must have at least one child 12 years old or younger, be considered
at-risk for abuse (either through presence of dynamics common
in abusive families or the presence of substantiated abuse
or neglect), and be willing to participate in services. Families
are served without regard to ethnicity, sex, number of children,
or age of the parent(s).
Magnitude and severity of problems and needs
the program addresses. An estimated 903,000 children were
confirmed as victims of abuse and neglect in the U.S. in 2001
(Child Maltreatment 2001, 2003, pg. 21). Of these confirmed
victims, only 58% received post-investigation services (Child
Maltreatment 2001, 2003, pg. 60). Within 6 months of the initial
substantiated or indicated maltreatment, 8.9% of abuse or
neglect victims suffer a recurrence of abuse (Child Maltreatment
2001, 2003, pg. 24). Approaches that not only treat abusive
families but also reduce further substantiations are of the
utmost importance. Parent aides are major partners in serving
at-risk and abusive families with an estimated 600 programs
treating over 59,200 families per year (Bryan, 1993). Parent
aide programs currently receive over 69% of their referrals
from DSS and serve all risk levels of families.
Geographic location, context and services
provided. Parent aide programs are located across the U.S.
The parent aide’s involvement is family-centered, empowerment-based,
and ecological in nature with services provided in the home.
Since parent aides are often volunteers or paraprofessionals
employed by a non-profit agency, they are typically able to
form non-threatening strength-based relationships with families.
The initial phase of involvement is labeled the “Dependency”
phase (duration: 3-6 months) in which relationship is built
with the family and concrete help is offered (e.g. food, housing,
childcare). The affective is emphasized to assist parents
in articulating feelings, problem-solving skills are taught
and emotional and crisis supports are offered. The second
phase is the “Interdependent” phase that addresses
more cognitive issues and begins the development of informal
and formal social supports. Parenting skills are discussed
and modeled within the framework of the family’s daily
environment. Over an average of 9 – 12 months, the family
progresses to the third phase of “Achievement”,
that is marked by a measurable level of stability, improved
parenting, and safe environments for children.
Conceptual, theoretical and/or practice
basis underpinning the program. Effectiveness is measured
based on four goals: child safety, problem-solving skills,
parenting skills and social support. Black et al in “Risk
Factors for Child Physical Abuse” (2001) identifies
inability to problem-solve as one of the key factors contributing
to abuse. Kolko in “Child Physical Abuse” (APSAC,
2001) identifies that “negative perceptions of children
and inconsistent child-rearing practices” contribute
to abuse. Barber & Delfabbro looked specifically at parenting
skills (2000) as an effective way to identify abuse/neglect
in families. Increased support has been linked to maintaining
improvements of intervention. Garbarino keyed in on social
support as a major factor in his influential book An Ecological
Approach to Child Maltreatment (1981). Many researchers have
continued to stress its importance including the numerous
studies by Jim Gaudin and Polansky in connecting social support
to neglect.This evaluation has incorporated testing and observation
tools in order to measure these outcome goals with families.
Description of the program to be evaluated.
A logic model demonstrates linkages. The evaluation will focus
on families receiving parent aide services through The Exchange
Club Child Abuse Prevention Center of N.C., Inc. known as
SCAN (Stop Child Abuse Now). Founded in 1981, SCAN was the
fourth center established as a part of the NECF network. The
Executive Director has guided its growth since 1981. SCAN
is located in the center of the state, provides services to
urban and rural families in Northwest N.C., has a history
of consistent adherence to the parent aide model, uses both
volunteers and paraprofessionals in to serve a racially diverse
population, has a sophisticated quality assurance policy,
and is credentialed at the highest level in Standards of Operation
and Practice within the NECF network. Please refer to Criterion
2, Question 3 for more information.
3.
Literature Summary.
A response to child abuse must include a
continuum of services. Home visiting has surfaced as one of
the preeminent strategies for primary, secondary or tertiary
prevention. Parent aide programs are providing intervention
in all levels of prevention. Parent aide and home visiting
programs are cited as exemplary, best practice or promising
programs from many facets of the research community. Family
Support America (2001) cites the importance of home visiting
from the angle of family support as intervention. Guterman
emphasizes the beneficial effect of home visitors for primary
and secondary prevention in his recent book Stopping Child
Maltreatment Before it Starts: Emerging Horizons in Early
Home Visitation Services (2001). From preventing out of home
placements perspective, Kelly concludes it is “preferable”
to have home-based programs (2000). There are many more citations
that emphasize the importance of home visitation and parent
aide intervention as a best practice or promising approach
and all conclude that there has not been enough stringent
research.
4.
Evidence supporting evaluation.
There have been several evaluations of Parent
Aide programs, but no randomized studies that have reviewed
a substantial sample. Anne Cohn, in a study by the Berkeley
Planning Associates in 1977 of 11 three-year child abuse and
neglect service programs looked at the use of Parent Aides
in the SCAN of Arkansas program (total sample was 1,724 of
which 207 were served by Parent Aides). In studying the reduced
“propensity for different types of maltreatment”
she reported that “In multivariate analyses, it appears
that certain treatment mixes – notable the lay service
model[Parent Aide]- remains the most effective variable in
explaining outcome. This is to say that when clients receive
the lay service model, irrespective of most of the case handling
or management techniques used, they are more likely to improve
while in treatment.” The study also analyzed cost per
family served with successful outcomes discovering the Parent
Aide intervention cost $2,590 per successful family which
was over $1,500 less than the next costliest intervention.
Due to the short period of studying the families, no follow-up
beyond the period served, and no control group the authors
found this study “suggestive of directions … programs
might take” and encouraged further evaluation. Krell,
Richardson et.al. ("Parent Aides as Providers of Secondary
Preventive Services- An Assessment" 1982) studied 58
families after two years of program operation. However, only
6 had terminated as planned. One third were reported for abuse
or neglect while being assisted. Hornick (1986) studied 27
families and a comparison group. After 1 year 74% of the lay
therapy subjects were still involved with services compared
to 50% of the families working with social workers. He concluded
this ability to maintain contact was an important result.
In addition, families served by lay therapists as compared
to non-lay therapist visited families showed significant differences
in the ability to empathize with the child and express pleasure
when with the child as compared to the control group. Miller
et al. (1985) studied 37 cases which had completed within
a two year period. The study only rated goals achieved by
their internal standards. It was concluded that the Parent
Aide program was efficient in spending 4.6 months in service
to a family at a cost of $1096 per year (1982). Carroll and
Reich studied 22 families (1978) after 2 years of programming
and found that 85% of the mothers were not re-reported for
abuse or neglect. Cameron and Rothery evaluated a Parent Aide
program in Hamilton, Ontario (1985). However the evaluation
was of 16 families with a comparison group of 28 over a short
period of time. They concluded that “considering the
relatively small size of the Parent Aide Program sample [n=16],
these patterns [i.e. the results reported] are merely suggestive
and not conclusive”. They did identify the cost per
case and found that over a seven month service period the
cost was $1,030.93 (in 1984 Canadian Dollars). Janes evaluated
a Parent Aide program in Idaho treating child neglect and
found that 75% of the parents showed increased awareness of
parenting and 75% demonstrated less neglect. Over the two
year study of 53 families 7.5% of families were re-reported
to the Department of Health and Welfare for neglect or abuse
issues. Their study showed a cost of $1,790 per family. They
concluded that “ …the use of the parent aides
was one of the major factors for success of this program.
Our parent aides provided to the parent, perhaps, for the
first time, a true support person.” (1988). Black, M.M.
et al (1994) studied home visitation with drug abusing women
with a sample of 31 reporting that infant cognitive scores
were higher and medical appointments more frequent but differences
declined at 12 months. Whipple and Wilson (1996) studied 34
families with no control group and found “significant
reductions in parental depression and stress”. The results
of Parent Aide program evaluations is limited and inconclusive
with few randomized studies and none with significant samples.
There are no published studies of variables which may cause
success or failure in parent aide intervention and there have
been no recent cost studies.
Cameron and Vanderwoerd in a section in
their 1997 book Protecting Children and Supporting Families
conclude after their review of Parent Aide evaluation attempts
that “No rigorous evaluations were found in this review.
The results from the small number of weak-design studies reviewed
generally showed few markedly superior benefit for families
from their involvements with Parent Aides. The literature
provided limited insights into how Parent Aides or para-professional
home visitors should be used to support disadvantaged families
or the kinds of benefits for families that can be reasonably
expected from involvement with Parent Aides.” Yet these
authors as several of the studies above have suggested, find
the Parent Aide model “… to be significantly different
and useful approach to the protection of children….”[1]
They further found that using volunteers in these roles [Parent
Aide] “…reduced child maltreatment and child placements
as well as produced improvement in parenting behaviors.”[2]
The review of the literature seems to clearly suggest that
Parent Aide programs have a descriptive positive effect on
child abuse and the overwhelming recommendation of a variety
of authors is that evaluation is needed. This proposal responds
to this important need with a significant sample and strong
randomized design with major investigators.
5.
Conceptual framework for evaluation.
This evaluation of Parent Aide services
measures effectiveness based on four researched goals of intervention:
child safety, improved problem-solving skills, improved parenting
skills and increased social support. Black, et al in “Risk
Factors for Child Physical Abuse” (2001) identifies
problem-solving as one of the key factors contributing to
abuse. Stress is a major inhibitor of effective problem-solving.
Parenting Skills are key to appropriate expectations, understanding
child behavior, and having the skill to interact with one’s
child successfully and effectively. Kolko (2001) identifies
that “negative perceptions of children and inconsistent
child-rearing practices” contribute to abuse. Barber
& Delfabbro looked specifically at parenting skills (2000)
as an effective way to identify abuse/neglect in families.
Increased social support has been linked
to maintaining improvements of intervention. Garbarino keyed
in on social support as a major factor in his influential
book An Ecological Approach to Child Maltreatment (1981).
Many researchers have continued to stress its importance including
research such as Crouch, Milner, & Thomsen in “Childhood
Physical Abuse, Early Social Support, and Risk for Maltreatment:
Current Social Support as a Mediator of Risk for Child Physical
Abuse”(2001) where in a large study they identified
perceptions of social support as related to higher risk for
abuse.This evaluation has incorporated testing and observation
tools in order to measure these outcome goals with families.
See logic model for more information.
Strengths. NECF network is over 24 years
old; consistency of training and program implementation; site
to be evaluated is 22 years old; evaluation will use multi-variate
analyses of data (see Criterion 2, Question 2); NECF has established
structure to assist in program replication; strong experimental
design of evaluation.
Limitations. Three-year time frame does
not allow for long-term tracking and observation of families
and limits the number of families that can be evaluated; demographic
variations, differences in child abuse laws.
6.
Contextual factors affecting implementation; opportunities
and barriers.
Factors hindering project implementation
How issue to be addressed
DSS movement to Multiple Response System (MRS).
Winston-Salem Center (SCAN) has been actively involved in
MRS implementation. Other areas adopting MRS have experienced
either stability or increase in referrals to outside agencies.
Loss of site’s funding base.
SCAN has been established since 1981 and has a diversified
funding base. Loss of any one source of funding should not
significantly impact services.
Inconsistency of referrals from community and DSS in particular
(due to DSS staff turnover).
SCAN has a close working relationship with its local DSS and
will continue to provide on-going education regarding referral
process to new DSS workers. SCAN has an Outreach Policy for
education of community referral sources.
Sample attrition.
Dr. Guterman has a study engagement protocol which successfully
enrolls and retains 95%-100% of families receiving services
into intervention studies (Guterman, 2003). SCAN has an 85%
rate of retention on cases served. Case management services
will be provided to families on the wait list group using
a specific protocol.
Barriers to project
How issue to be addressed
Limited time frame of 3 years does not allow for long-term
follow-up of sample and control families nor allows for sufficient
analyses of data.
Additional funding will be sought to allow for long-term follow-up
with families participating in the evaluation.
A no-cost extension will be requested to
allow for more complete analyses of collected data.
Small sample size.
Proposed evaluation is carefully structured to maintain sample
integrity (see Criterion 2).
Control group issues (e.g. removal of child from the home;
crisis of family) might necessitate removal from wait list
in order to ethically provide services.
Case management services following a rigorous protocol will
be provided to families on the wait list group to decrease
attrition due to removal of child or family crisis.
Factors facilitating project include: DSS
Movement to MRS, increase of referrals from DSS and community;
receipt of additional funding allowing for expansion of parent
aide services. Each of these factors could lead to an increase
in sample size for the project. Opportunities in project design
and implementation include: sample size increase, analyses
of existing data that has been collected on 672 families receiving
parent aide services. In addition, this evaluation could serve
as a spring-board for future studies of parent aide services
and would evaluate a promising program that, to date has not
been adequately studied.
7.Results
and benefits to be derived from evaluation.
There has not been a successfully conducted
random evaluation of parent aide programs to date to determine
if services are effective. Preliminary evidence indicates
that parent aide services are cost-effective, decrease child
abuse and neglect recurrence, and can be replicated within
a variety of communities. There are no published predictors
of success in parent aide services yet these services are
provided everyday by over 600 programs nationwide (Bryan,
1993). This evaluation will provide a scientific evaluation
of parent aide efficacy, will evaluate cost-effectiveness,
and identify predictors of success in program delivery to
improve practice.
8.
APPROACH
Timeline for implementation.
Within the first two months, the Site Coordinator/Liaison
(hereafter referred to as Liaison) will be trained for the
Winston-Salem study site. By month 3, a system will be in
place to carry out the randomization of eligible families
into either the treatment or control arm of the study. Simultaneously,
the Columbia research team will prepare a computer-assisted
interview and self-interviewing (“CASI”) protocol
to be administered on laptop computers during in-home interviews,
and drawn from the measurement protocol described. Columbia
staff will submit an IRB protocol. At this point, families
will be recruited into the study (month 2-3). The Liaison
will receive a list of parent aide eligible families and recruit
them into the study up to month 30. Each recruited family
will be randomly placed into the treatment or control group.
Once a family has been recruited into the study, the Liaison
will provide the essential contact information to the Data
Collector. The family will be contacted for a baseline interview
to occur prior to service delivery. The Data Collector will
track recruited families and contact them again at least 1
month prior to both the 6- and 12-month follow-up interviews.
All baseline interviews should be completed by month 26, all
6-month interviews by month 31, and all 12-month interviews
by month 36. Two data files will be constructed and completed
(interim file by month 18; final file by month 36). Accordingly,
interim data analyses will be conducted at Columbia during
months 18 and 19 to review the data quality. Once the final
data file has been received, final data analyses will be conducted.
A final summary report and relevant manuscripts will be prepared
based on the findings. Attrition of families poses the greatest
threat to meeting proposed milestones of this study. We will
build in an intensive set of procedures to maximize engagement
and retention (see Item 3).
2. Issues related to program evaluation in general, issues
for evaluating this program in this context
9.
Evaluation methodology.
A prospective experimental design will
be used to test the effects of a parent aide program designed
to improve four main outcomes in families: 1) child safety,
2) problem solving skills, 3) parenting skills, and 4) social
support. Five program sites serving northwest N.C. will participate
in this study. Referrals come from the following sources:
46% DSS; 28% Self-referral; 13% medical personnel; 4% schools;
9% other. Referral issues include: 40% abuse, 23% neglect,
10% abuse and neglect, 27% high risk. Other demographics of
referred families are as follows: 72% Whites, 28% African
American; 49% unemployed; 45% below high school, 35% high
school graduates, 13% some college/technical, 7% college graduate;
74% voluntary, 3% court suggested, 12% court ordered, 11%
other.
In order to be eligible for study participation,
a family must be referred to one of the five parent aide programs
cited above. Eligible families will be asked to meet with
the Liaison to conduct an initial eligibility review for parent
aide services and complete an informed consent interview (specifying
study nature, procedures, time commitment, remuneration for
participation, risks and benefits, confidentiality and its
limits, the voluntary nature of participation, and the protection
of data). Families will be screened out and referred to other
services if the parent: has an IQ of 60 or below; exhibits
a psychotic disorder and cannot be stabilized quickly; or
if the children have been removed from the home by DSS and
reunification is not a goal. Families that decline participation
will receive regular services with no consequences. Once families
provide informed consent, a baseline interview will immediately
be arranged with the Data Collector in the home, to be completed
prior to random assignment and any service delivery. The Liaison
will assign the family to either 1) a parent aide + case management
service condition (hereafter referred to as the “intervention
group”), or a wait list control (case management only)
service condition (hereafter referred to as the “control
group”), using a random assignment list generated using
a random number generator.
For the families assigned to the intervention
group, they will be matched to a parent aide and provided
in-home services that will also include referral and monitoring
of referrals to other resources in the local community. Parent
aides will be trained and supervised both individually and
in a group, and will follow a well-developed protocol using
the 25-page “Parent Aide Casework Manual” (revised
2003), first developed by George Bryan for the Winston-Salem
site. This manual has been adopted by other sites and is used
by NECF as a part of new director orientation to ensure program
consistency among network sites. According to this protocol,
an INA in conducted (please refer to Criterion 1, Question
1). After this initial session, 1-2 in-home visits are provided
each week (duration 1-1.5 hours). This level of service intensity
tapers after families are stabilized (3-6 months). We will
carefully monitor and collect data on service dosage and referrals
to other community resources in both the intervention and
control groups.
The control group condition will engage
families with a staff member who will provide case management,
referral, and monitoring by phone, and place families on a
6-month wait-list to receive parent aide services. Crisis
intervention services will also be provided if/when necessary.
This condition averts the ethical difficulty of withholding
services from at-risk families, as the Winston-Salem sites
already maintain wait lists of families who are provided case
management services (approximately 25% of those referred).
Currently, wait list families receive INA and 1 contact per
month. Case management services will be enhanced so that families
in both the intervention and control group conditions will
receive a greater degree of case management services (i.e.
bi-weekly phone or in-person contacts; referrals to other
services either internal or external to SCAN; and delivery
of requested information, e.g., literature on time-out). After
six months, those families randomly assigned to the wait list
control group will become eligible for a parent aide match,
and enter the intervention arm of the study. This will enable
us to observe differences in outcomes after six months of
services, and then to track any attenuation of those differences
at 12 months. We will block random assignment according to
the five sites, so that families receiving parent aides and
those on the control group will be evenly distributed.
This experimental design provides a maximum
degree of control of potential extraneous factors that might
otherwise confound our observation of intervention effects,
and simultaneously provides for all enrolled families a minimum
of case management services that are more intensive than those
presently provided. The randomization process is designed
to ensure that, overall, the characteristics of the intervention
and control groups are equivalent. We will conduct a baseline
pretest, as well as 6- and 12-month follow-up assessments
to evaluate the changes in the four primary outcomes of interest
previously mentioned that may be attributable to the receipt
of parent aide services (child safety, problem-solving skills,
parenting skills, and social support). Although some families
will receive more than 12 months of parent aide services,
this study design will be able to detect parent aide effects
at data points where differences are most likely to occur,
i.e. during the first 6 months of typically more intensive
support (compared to the control group who will not yet have
received any parent aide services), and at 12 months, at which
time services typically taper in intensity. If, as we hypothesize,
we see clear intervention effects, we will seek separate external
funding to continue to follow-up the cohort past the 12 month
follow-up period.
We considered enrolling families at other
parent aide program sites in other cities across the U. S.
to increase ethnic diversity of the sample, as we were concerned
with the generalizability of the findings. Given the funds
available for evaluation, the additional challenges that a
multi-site study brings that would harm the internal validity
of the study (e. g. differences in organization, service delivery
patterns, supervision, quality control mechanisms, and alternate
wait list conditions), we opted to focus this initial study
within the Winston-Salem sites. This enables us to execute
a study with a very high degree of internal validity from
which to assess the effectiveness of parent aides. Once we
garner clear outcome findings from this site, we will then
be in a position to extend this study to other sites.
The quality of this study relies heavily
on the retention of families for all three data collection
points to minimize biases that may be attributable to attrition
due to self-selection factors. One of the clear strengths
of the proposed study is that we have built in a number of
procedures that will diminish the likelihood that attrition
will either introduce such biases or harm the statistical
power of the study and sample size. Presently, 70 families
per year are matched to parent aides at the Winston-Salem
sites, and 36 are placed on a wait list for parent aide services.
We have chosen the Winston-Salem sites as they engage an average
of 78% of the families offered services per year, one of the
highest engagement rates of any Exchange-sponsored parent
aide program in the country, and comparatively, a very high
engagement rate for a largely involuntary population of CPS
referred cases. Although it might be expected that some families
will choose not to enroll in the study but still wish to receive
services, our study engagement protocols successfully enroll
and retain 95%-100% of families receiving services into our
intervention studies (Guterman, 2003). These procedures involve
not only remunerating families for their participation, and
explaining the purposes and benefits of studies, but also
engaging families in an active problem-solving discussion
to assist them in identifying potential facilitators and obstacles
to their participation (e. g. scheduling difficulties, unsympathetic
other family members, personal hesitations) as well as strategies
to overcome identified barriers (c. f. McKay, et al, 1998).
Interviewers also work with service staff as a bridge to engage
families, develop extensive contact sheets that are periodically
updated, make interim phone contacts, and collect data in
the homes at a time set by the study participant (Prinz, et
al., 2001).
We are developing procedures with the N.C.
DSS to increase the number of referrals to the study sites,
which will increase the number of families that will enter
the randomized trial and particularly increase the wait control
list. Estimating conservatively, we expect a minimum of 134
families per year to be referred for services (70 matched
to parent aide plus 64 on a wait list). Assuming a 78% engagement
rate, we expect to minimally enroll: (134 families referred
X 78% = 104 families enrolled per year); (104 families enrolled
per year X 2.0 years of enrollment = 208 families enrolled).
The Winston-Salem sites commendably retain 82% of those families
enrolled until cases close with goals attained, with services
lasting in length ranging from a few months up to 2 years,
57% of which last more than one year of service. Approximately
90% of enrolled families are retained during the first year
of service. Given this, we expect to retain: (208 enrolled
families X 92% (conservative estimate) = 191 families @ 6
mos. follow-up) and
(208 enrolled families X 87% (conservative
estimate) = 183 families @ 12 mos. follow-up). As presented
in the power analysis below, these sample sizes provide satisfactory
statistical power to detect significant differences between
intervention and control groups.
9. Variables
and Their Measurement.
Program sites are notable for the extensive
data already collected in case files of 672 families and we
will make use of these data in a variety of ways. In addition
to executing a retrospective baseline study to assess the
degree to which families enrolled in the prospective randomized
trial are similar, we will also use case records to minimize
data collection burden on families and staff members. Demographic
information that may identify predictors of success in parent
aide services will be abstracted by the Data Collector from
case files, and will include such factors as parent’s
education, members of the household, marital/family status,
ethnic backgrounds, ages, children in and out of the home
and their ages, household members, income and prior encounters
with child protective services. At follow-ups, we will also
abstract information to assess the intervention processes
(described below), including referrals to other community
services and resources.
At baseline, 6- and 12-months follow-up,
the Data Collector will interview mothers using Audio-Computer
Assisted Self-Interviewing technology (“A-CASI”)
technology to minimize self-report biases. For the proposed
study, we will interview mothers only, and not other family
members (e. g. fathers), so as to maximize homogeneity in
the sample. Given that the vast majority of direct service
recipients of parent aides are mothers and this study’s
modest sample size, collecting data from others (such as fathers
or grandmothers) will not yield large enough sub samples to
analyze these groups, and will introduce a host of additional
complexities when trying to interpret outcome findings. Further,
collecting data from mothers only is in line with our focused
query on the direct impact of parent aide services on the
most common service recipients. We will, however, collect
data on significant others via the Vaux social support scale
(below), and via our abstraction of demographic information
from case records.
Interviewers using A-CASI technology provide
laptop computers to study participants when sensitive or emotionally-laden
questions are posed (e.g. those concerning maltreatment).
Rather than asking such questions face-to-face, sensitive
questions are heard by participants privately on headphones,
and their answers are directly keyed into the laptop computer,
maximizing the privacy and confidentiality of responses. Studies
have shown that A-CASI technology substantially minimizes
self-report biases, enhancing data validity, particularly
around questions concerning violent and illicit behavior (Turner,
et al., 1998). Simultaneously, direct entry of data onto laptop
computers eliminates the steps of paper and pencil data collection
and then transfers to a computer file, saving the costs of
manual data entry and error checking, and eliminating an important
source of data error.
The data collection battery will assess
changes over time in relation to the four primary objectives
of the parent aide program. In order to enhance study retention
and to compensate family members for their time and effort,
they will be provided $25 for every interview/data point they
complete (baseline, 6- and 12-months). The assessment battery
is expected to last 1 ¼ hours in the home, inclusive
of the Data Collector’s observation period. Although
the Data Collector will receive contact information to locate
and schedule interviews, they will be “blinded”
as to group assignment (intervention versus control group).
No questions will be on the data collection battery that will
reveal participants’ group status. Study outcomes will
be assessed in the following ways:
Official CPS Reports. Under N.C. Juvenile
Statutes, DSS may share information freely to treatment resources.
Victoria Worden, Program Manager of Child Protective Services
in Forsyth County, N.C. has submitted a letter indicating
that their department will negotiate to provide needed information
if this grant is awarded (July 23, 2003). A copy of this letter
can be provided upon request. These data will be coded only
after we have removed identifiers from the dataset to assure
confidentiality, in relation to occurrences of abuse and neglect.
Data will be coded for type and severity of the report, allowing
us to assess change in maltreatment reports over time. Reports
to protective services systems, while key in establishing
the efficacy of parent aides, also hold important limitations
in their utility for a randomized trial such as the one proposed.
The increased monitoring in the home that accompanies parent
aides may serve to heighten the likelihood that families will
be reported to protective services, thereby potentially confounding
and masking intervention effects. Further, imprecise definitions
of child abuse and especially child neglect (e. g. Zuravin,
1999), as well as differing judgments across professionals
about the presence of abuse and neglect (e. g. Rose and Meezan,
1993) do raise questions when attempting to draw conclusive
inferences based on official reports of maltreatment alone.
Given the primary outcomes of this study and the limitations
in official CPS data, we will triangulate our data collection
from CPS with self-report data from parents, as well as with
in-home observation on parenting skill, capacity, and home
safety, assessing varied forms of child neglect, physical
abuse, emotional abuse, and sexual abuse. The following instruments
will be used to directly assess child maltreatment and parenting
skill in the home:
Parent-Child Conflict Tactics Scale (Straus,
et al., 1998): used across a variety of cultures; has established
sound reliability and validity; includes nonviolent discipline,
psychological aggression, and physical assault scales, as
well as supplementary scales for child neglect and sexual
abuse; is behaviorally specific which will enable assessing
change over time, and minimize self-report biases. We will
adapt the scale to assess prior 6-month (rather than 1-year)
time periods, consistent with our data collection intervals.
Child Well-Being Scales (CWBS) (Magura
& Moses, 1986): assesses 43 dimensions of parenting performance,
familial capacities, child performance and child capacities,
across a wide age range of children; has shown strong psychometric
properties including criterion and content validity and interrater
reliability.
Childhood Level of Living Scales (CLLS)
(Polansky et al., 1978): consists of 99 dichotomously scored
items related to the quality of physical, emotional, and cognitive
care provided by the parent; has reported sound psychometric
properties. We will adapt the scale for a wider variation
in ages of children, as the original scale was developed for
children 4-7 years old.
Parental Problem-Solving Measure (PPSM)
(Hansen et al., 1989): assesses the problem-solving skills
of parents, assesses child behavior management, anger/stress
control, financial, child care resources and interpersonal
problem-solving skills using vignettes and open-ended questions;
has reported strong psychometric properties including internal
consistency, and content, criterion and construct validity.
Vaux Social Support Appraisals (SS-A) Scale
(Vaux, et al., 1986): assesses parents’ beliefs that
they are supported by and involved with family, friends, and
others; has demonstrated strong psychometric properties including
internal consistency, concurrent, convergent, and divergent
validity; has shown comparatively strong predictive capacity
when compared against other measures of social support networks.
Given that parent aides aim to reduce child
maltreatment risk by reducing parental stress, and by increasing
their sense of empowerment, we will also include in the measurement
instrument key measures assessing these potential mediating
variables:
Parenting Stress Index—Short Form
(PSI) (Abidin, 1995): assesses felt stresses in the parenting
role; has shown excellent psychometric properties; been found
to closely predict child maltreatment risk; been successfully
used in maltreatment prevention evaluation studies.
Pearlin-Schooler Mastery Scale (PSM) (Pearlin
& Schooler, 1978): assesses parents’ personal sense
of control over life circumstances and is a close proxy assessing
the degree to which parents feel greater empowerment and mastery
in their lives; has shown excellent psychometric properties
including internal consistency, construct, predictive, and
discriminate validity; has been successfully used in child
maltreatment preventive interventions, which have shown personal
control, to both moderate the effectiveness of early maltreatment
prevention services (e. g. Olds, et al., 1986), and to be
a positive outcome of such services, as assessed by this scale
(Kitzman, et al., 1997).
10. Process
Variables and their Measurement:
Specific process factors must also be understood
in the role that they play in predicting specific outcomes
observed. In addition to assessing outcomes in connection
with the receipt of parent aide services, we will also examine
a set of process variables that will help in establishing
predictors of parents’ participation in and benefit
from services, the factors that both facilitate and hinder
parents engaging in and benefiting from parent aide services.
These variables and their measurement, which will be completed
only on the parents’ randomly assigned to receive a
parent aide, will include:
Parents’ attachment to parent aide/case
manager: measured by an adapted version of the Barnard Attachment
to Home Visitor Scale (1998), which assesses the strength
of the working relationship between a parent and their parent
aide or case manager. This will be completed in the A-CASI
section of the interview, so as to prevent the data collector
from learning the group status of the family being interviewed.
Parent satisfaction questionnaire: assesses
the degree to which parents are satisfied with their parent
aide, would recommend their parent aide to others, feel they
are benefiting from their services, and other process evaluation
questions.
Engagement and retention: to be measured
by the proportion of parents who are linked with a parent
aide that then hold at least a first meeting with their parent
aide where services are provided (“engagement”);
and by the proportion of parents who receive services until
case goals are attained or until services are closed for mutually
agreed parent aide and parental (rather than parents’
sole) reasons (“retention”). Length of such services
will be coded as well.
Service dosage: will be measured by the
number of parent aide in-person sessions held with the family,
and the duration of those sessions.
11. Cost Variables
and Their Measurement.
We will gather information from administrative
records on the costs of service (personnel and non-personnel
expenses associated with the parent aide service), and calculate
total costs of service per case by using a multiplier from
the duration and dosage of services. We will also gather information
from both the intervention and control groups on social, health,
and other formal services and benefits they have received
from the point of referral, collecting these data at each
outcome point. We will follow along the “CPPOA”
model (cost-procedure-process-outcome analysis) of cost effectiveness
(Yates, 1996) to examine links between material costs, intervention
processes, client mediating factors and observed outcomes
as assessed above. Although for this three year grant, we
will be unable to follow-up families over a longer post-service
follow-up period, we will nonetheless track service use, encounters
with authorities, and use of public services and entitlements,
which can be monitored for a cost-benefit analysis. We will
consult with Prof. Irwin Garfinkel (Columbia University School
of Social Work), who holds specific expertise in cost-benefit
analysis of social programs to assist us in monitoring the
less tangible outcomes, to carry out a cost-benefit analysis
on the sample in this study.
12. Data Analysis
Plan.
Data analysis will be conducted to examine
the effects of the parent aide program. This assessment will
be made by comparing the dependent variables across group
(intervention vs. control) and over time (pre vs. post-test)
using multiple linear regression. This type of analysis, entering
intervention vs. control as a dichotomous predictor in regression
equations, will also allow for the statistical control of
multiple covariates, including demographics and potential
moderating and mediating variables, and maximize statistical
power. This strategy allows us to examine for dosage and attrition
effects on outcomes. Data will be analyzed according to an
“intent to treat” model, the most scientifically
stringent strategy, where all families that are randomized
into the study, regardless of whether they are retained in
the services and study will be included in the analysis. Data
will also be analyzed according to a “per protocol”
basis, where only families receiving services until the case
goals are attained, are included in the analysis, and findings
across analytic strategies will be compared. If our attrition
rates become unexpectedly larger than planned, we will conduct
attrition analysis to examine for attrition biases, and we
will be prepared to employ multiple imputation strategies,
using such software packages as AMELIA (King et al., 1999)
to minimize these biases and extend statistical power in the
data analysis.
In examining the statistical power of this
data analytic strategy, we assume a mild to modest effect
size (d = .4). We have based this conservatively on preliminary
evidence derived from such studies as Berkeley Planning Associates
(1997) and Whipple & Wilson (1996) and Winston-Salem’s
data compilation (1997). Winston-Salem’s data compilation
(1997) indicates a 2-3% maltreatment recurrence rate at immediate
follow-up with services averaging approximately 12 months,
in contrast to an 8.5% recurrence rate after 6 months in the
State of North Carolina of 8.5% (U. S. DHHS, 2003), and an
average pretest to posttest gain on worker reported goal attainment
scales of 2 points (40%) on a 5 point Likert-type scale. With
an N of 183 families at 12 month follow-up (approximately
191 at 6 months), and a using a 1-sided probability test and
significance level of p < 0.05, the resultant power at
the 12 month follow-up point is .85 (.87 at 6 months), meaningfully
above the .80 convention of acceptable statistical power (providing
us an 85% probability at 12 months and an 87% probability
at 6 months, conservatively estimated, of finding statistically
significant differences when they empirically exist in the
larger population). If, in the unlikely event that the sample
size does not reach at least 180 using 30 months of enrollment
at the Winston-Salem sites, we will extend the enrollment
window for 6 additional months to the close of the grant period,
increasing the sample size by 30-35 additional families, and
we will then conduct data analysis and write-ups during a
six month no-cost extension period to the grant.
4. Relationship between applicant and the
program to be evaluated. NECF will contract with Dr. Neil
Guterman (Columbia University School of Social Work) as outside
evaluator. Dr. Guterman is a leading researcher within the
prevention field and has already worked extensively with the
Program Site and NECF on this proposal. A positive working
relationship will be maintained through communication via
regular teleconferences, at least yearly in-person meetings,
and a written description of roles and expectations. To initiate
this process, a Letter of Agreement has been signed by all
parties and is enclosed as a part of the application packet.
Roles, responsibilities, and key staff are further defined
in Criterion 4, question 10.
5. Plan for ensuring the evaluation will
be culturally sensitive to the target population. The battery
of outcome measures (detailed in Question 2 above) have been
selected for their psychometric strengths and efficiency (to
minimize burden on study participants). All measures have
been successfully employed with both African American and
White study participants (who constitute the vast majority
of expected participants in this study), with many measures
having reported strong cross-cultural validity across an array
of ethnic groups (e. g. the Conflict Tactics Scale and the
Parenting Stress Index). The Data Collector will travel to
the homes of participants to maximize their involvement, to
enable observational data collection regarding home safety.
All study participants speak fluent English, so there will
be no need to prepare the data collection battery in another
language."
6. Plan for identifying program components
and strategies most closely linked to desired results. The
following documentation will be used to predict which kinds
of support provided are associated with which kinds of desired
outcomes. These components will assist in isolating "components"
of service and statistically examine for links with observed
positive outcomes. They will also provide a means to "check
the fidelity" of the intervention These documents include:
referral form; INA; narrative of each family contact detailing
provided services, parent education and dosage; supervision
notes; correspondence; evaluations; treatment plans; termination
summary; pre- & post-evaluation of progress toward four
long range goals; data sheet collecting 125 different pieces
of information on each family, e.g. income, abuse).
7.Existing data to be used in the evaluation.
SCAN has compiled information in an Access database on families
served since 1981. Reported data in this compilation includes
Race, Occupation, Education, Religious Participation, Referral
Type, Case Type (i.e. type of abuse/neglect), Referral Source,
Length of Service, Volunteer Parent aide Service, Severity
of Abuse or Neglect, Type of Injury [to the child], Special
Characteristics of the Victims. Existing case and administrative
files will be reviewed retrospectively to mine information
related to program/participant characteristics and cost-benefit
analysis.
8.New data that will be collected. 1) DSS
records will be examined for abuse recurrence and out of home
placement incidences; 2) cost-study; 3) randomized trial data;
4) results from the various measurements indicated in Question
2.
9. Plan for ensuring confidentially, informed
consent and IRB review. Approval is being sought for this
study and will be received by Columbia University’s
Institutional Review Board (“IRB”) prior to the
enrollment of any subjects and collection of data for this
evaluation. We will adhere to Columbia IRB procedures to assure
fully informed consent, awareness of risks and benefits of
participation, assurances of confidentiality and its limitations,
protection of data, and the voluntary nature of participation.
14. Roles and
responsibilities of each partner &/or agency.
NECF (Applicant) shall: provide the fiscal
management of the project; ensure that program sites are in
compliance with National program standards, develop and execute
subcontracts; coordinate travel and meeting arrangements;
handle all receipts and bills; and provide the financial reporting
to the Administration on Children, Youth and Families.
Dr. Neil Guterman, Columbia University School
of Social Work, New York (Principal Investigator) shall: coordinate
activities of the evaluation; ensure the timely collection
of data, its conversion to other software, and its analysis;
ensure compliance with all Departmental regulations and procedures
pertaining to confidentiality and careful handling of data,
obtain informed consent from participants, and submit to an
Institutional Review Board (IRB) review. Dr. Catherine Taylor,
Columbia University, will act as Project Manager and shall:
be responsible for hiring, training; and supervising data
collectors; shall work in conjunction with Program Site and
P.I. to ensure that the evaluation proceeds within budget
and time constraints.
SCAN Center Executive Director shall: act
as Project Director and will coordinate activities of the
evaluation; be responsible for insuring coordination of data
collection; facilitate staff-evaluator discussions regarding
data analysis; be the liaison with NPAN and with the Administration
on Children, Youth and Families; be responsible for reports
that are not financial in nature; write the final report with
major contributions by the P.I., and be involved in dissemination
of the information.
Site Coordinator/Liaison: will act as the
local liaison in the grant and shall ensure that local staff
are involved in feedback to the Principal Investigator (P.I.);
facilitate access to client and/or client files by the P.I.;
and be involved in the Grant’s Team Meetings. The Liaison
is an employee of SCAN, Cynthia Napoleon Hanger.
15. 11.Products that will be developed during
the evaluation. Semi-annual reports; final reports; cost analysis;
process analyses; catalogue of characteristics that predict
success of parent aide intervention; journal articles, conference
presentations, Internet applications (postings to list serves,
web sites); a data collection engagement and retention protocol
and Audio-CASI data collection module for use on a laptop
computer. The audience will be child maltreatment practitioners
and policymakers.
12. Plan for summarizing, reporting, and
disseminating findings. Our research team is well versed in
dissemination of findings through scholarly and professional
venues. Dr. Guterman’s roles as Editor in Charge of
Prevention for the APSAC Advisor (of the American Professional
Society on the Abuse of Children) and an active ad hoc reviewer
for both Child Abuse and Neglect and Child Maltreatment journals
will assure timely and high profile dissemination of such
findings to this highly interdisciplinary audience of child
maltreatment professionals nationally. In addition, results
shall be shared through: Presentations at Regional, State,
National and International Conferences of Child Abuse and
Neglect; presentation at NPAN conference and NECF Symposium;
List Serve notices (NPAN, Child Abuse Prevention Network List
Serve--Cornell University); direct mail pieces to programs
included in the National Parent Aide Directory. Findings will
be used by NECF to improve training and National Standards
of Operation and Practice.
13.
CRITERION 3: ORGANIZATIONAL PROFILES
1. Applicant’s capabilities and experience
relative to this project. NECF has over 20 years of administering
and coordinating national efforts and employs full-time permanent
professional and clerical staff necessary to maintain the
infrastructure of a national non-profit child abuse prevention
organization. Infrastructure exists to manage this evaluation
grant. Two field-based professional staff provide technical
support and training to the network of centers. Additional
support is available through the Project Director’s
office. Winston-Salem SCAN Center (Program Site): The Executive
Director is George M. Bryan Jr. (Project Director) who has
over 24 years of experience in the field of child abuse and
neglect, currently manages child abuse prevention and treatment
programs throughout N.C., has been involved in several evaluations
of national scope (AARP study), and has administrative and
support staff to help in the coordination. The evaluation
research will be implemented via a subcontract to research
team directed by Professor Neil B. Guterman at the Columbia
University School of Social Work. Dr. Guterman holds over
12 years experience conducting research and evaluation studies
specifically in child maltreatment prevention, and is the
Principal or Co-Principal Investigator on a number of studies
ongoing in this field. He has recently received a planning
grant to initiate an endowed child maltreatment prevention
research program at Columbia University under which several
interrelated maltreatment prevention research initiatives
will be organized.
2. Applicant’s capabilities and experience
relative to child abuse prevention programs. NECF has provided
leadership and direction for its child abuse prevention network
since 1979. The role of NECF has been to provide training,
on-going technical support and quality assurance to the boards
of directors, volunteers, and professional staffs of the local
centers. All Exchange Club Child Abuse Prevention Centers
are incorporated in their respective states as private, non-profit
501(c) 3 organizations and are governed by volunteer boards
of directors. Each center is run by a professional director
and a variety of staff is utilized to provide parent aide
home visitation services and other prevention services as
appropriate to the local community. Over the course of the
past 20 years, NECF’s Child Abuse Prevention network
has provided home visitation services to 150,000 families
at-risk of abuse. Compliance with National Standards of Operation
and Practice are required of centers within the NECF network.
A national list serve is also maintained by NECF for the sharing
of research and ideas important to parent aide programs.
3. Knowledge, experience, and capabilities
that the key project staff brings to the project.
List of organization/consultants involved
with project and information on their roles, responsibilities
& time commitment as well as background and qualifications.
George Mezinko, Executive Director of NECF (Applicant) shall:
provide fiscal management; ensure that SCAN is in compliance
with National program standards, develop and execute subcontracts;
coordinate travel and meeting arrangements; handle all receipts
and bills; provide financial reporting to the Administration
on Children, Youth and Families. Qualifications: Executive
Director of NECF since 1985; M.Ed. in Educational Psychology;
extensive experience in programmatic, administrative and fiscal
management of a National organization. Anticipated time commitment
is 10%.
George Bryan, Executive Director of Winston-Salem,
NC Center (Program Site) as Project Director shall: coordinate
evaluation activities; be responsible for insuring coordination
of data collection; facilitate staff – evaluator discussions
regarding data analysis; be the liaison with NPAN and the
Administration on Children, Youth and Families; be responsible
for reports that are not financial in nature; write the final
report with major contributions by other key staff; participate
in annual grantee meetings to be held in Washington, D.C.;
and be involved in dissemination of the information. Qualifications:
Executive Director of Winston-Salem Center since 1981; M.Ed.
in Community Ministry; over 24 years of experience in the
field of child abuse and neglect; currently manages child
abuse prevention and treatment programs throughout N.C.; previous
and current involvement in evaluations of national scope;
founding board member of NPAN; extensive speaking and training
experience. Anticipated time commitment is 15%.
Dr. Neil Guterman, Associate Professor,
Columbia University School of Social Work (Principal Investigator)
shall: be responsible for the study’s strategic direction,
oversee its implementation, supervise the research team, and
coordinate with Mr. Bryan and the study sites. He will be
a primary author of peer review articles summarizing findings
from the study, and help assure the visibility of the findings
of the study on a national scale. He will attend the annual
grantee meetings held in Washington D. C. Qualifications:
Associate Professor and Doctoral Program Chair at Columbia
University School of Social Work; 12 years clinical experience
in social services; extensive experience of research and publication;
author of Stopping Child Maltreatment Before It Starts: Emerging
Horizons in Early Home Visitation Services (Sage Publications,
2001); extensive experience in conference presentations, teaching,
and training. Anticipated time commitment is 20%.
Dr. Catherine Taylor, Columbia School of
Social Work, (Project Manager) shall: act under the direct
supervision of Dr. Guterman; manage the implementation aspects
of the study (e.g. hiring and training study personnel); set
up randomization and data collection procedures; set up the
measurement battery and CASI modules on laptop; working closely
with the study site liaison; monitor the quality of the data;
troubleshoot to assure maximum participation in the study;
serve as primary data analyst; attend the annual grantee meetings
held in Washington D. C.; and collaborate on report and article
writing. Qualifications include: 4 years clinical experience;
significant research experience; presently serving in a post-doctoral
position with Prof. Guterman, assisting in launching an endowed
child maltreatment prevention research program at Columbia
University. Anticipated time commitment is 80%.
Cynthia Napoleon-Hanger, Deputy Director
of Winston-Salem, NC Center (Program Site) as Site Coordinator/Liaison
shall: act as the local liaison in the grant, ensure that
local staff are involved in feedback to the Principal Investigator
(P.I.); oversee the implementation of the study design at
the five Winston-Salem sites, manage the study-service delivery
interface; ensure that the ethical standards of the evaluation
are carried forth. This team member will work on-site, closely
collaborating with Dr. Taylor, to oversee the integrity of
referral and random assignment process, to conduct informed
consent interviews, intensively track families to minimize
study attrition, and liaison with service personnel, minimizing
staff burden. Qualifications: Deputy Director of SCAN since
1991; M.Ed in Human Services; over 30 years experience in
social services and child abuse prevention; extensive training
experience. Anticipated time commitment is 20%.
Data Collector (TBD) will be hired to locate
families, engage and make arrangements to interview them,
conducting all data collection sessions in-home with families.
We will seek to hire one data collector for consistency and
to maximize the likelihood that families will form a connection
with this person to minimize dropout in study participation.
This person will be trained and supervised by Dr. Taylor with
frequent conference calls, regular Internet contact. Training
will take place in-person. Anticipated time commitment is
75%.
4. Relationship between the proposed project
and other work planned, anticipated or underway with Federal
assistance. Non-applicable.
5. Management plan. The evaluator/ contractor
and site personnel will be paid by monthly reimbursement for
work performed. Work shall be reviewed each month by the Project
Director with payment made by the NECF. Project Director is
also the Executive Director of the Program Site and has the
ability to allocate resources, provide training and encourage
the program site’s coordination with the P.I.
6. Role of the author of this proposal in
the implementation of the proposed project. George Bryan,
Jr. is the Project Director. Dr. Neil Guterman is the Principal
Investigator.
CRITERION
4: BUDGET AND BUDGET JUSTIFICATION
1. Explain why the costs of the proposed
project are reasonable. We propose three different aspects
of evaluating the parent aide program—cost analyses,
randomized study, and retrospective analyses. The randomized
study is especially costly given the administration of a large
number of measurement tools (see Criterion 2, Question 2)
and observation within the home. In addition, an outside evaluator
will be hired to maintain objectivity. After preparing the
budget, it is clear that large amounts are in-kind contributions,
e.g. clerical support, past data collection at program site,
oversight of services by program sites. Given the tight timelines,
we would request a no-cost extension to allow for complete
analyses of the data.
2. Describe the fiscal control and accounting
procedures to be used. A part-time accountant produces monthly
financial statements. The fiscal year is July 1 through June
30 and an annual certified audit is conducted each year. Accounting
policies comply with Generally Accepting Accounting Principles.
All receipts and bills will be forwarded to the Executive
Director of NECF who will verify the expense against the grant
budget. Quarterly statements will be issued to the Project
Director and the Executive Director of the NECF to insure
the budget constraints are met.
3. Allocate sufficient funds in the budget
to provide for attendance at an annual grantee meeting.
Funds have been allocated for Project Director,
Principal Investigator, Site Coordinator/Liasion and Project
Manager to travel to annual grantee meeting.
16.
Bibliography
Abidin, R. R. (1995). Parenting Stress Index,
3rd Edition. Professional Manual. Odessa, FL, Psychological
Assessment Resources, Inc.
American Association of Retired Persons.
(1987). Results of a nationwide survey of parent aide programs.
Washington, DC.: AARP Social Outreach and
Support Program Department.
Barnard, K. (1998). Developing, implementing,
and documenting interventions with parents and young children,
Zero to Three, Feb/March, 23-29.
Berkeley Planning Associates & Cohn,
A.H. (1974-1977). Child abuse and neglect treatment programs:
Final report and summary of finding from
the evaluation of the joint OCD/SRS national demonstration
program in child abuse and neglect. California.
Bryan, G. (1993). National parent aide survey.
National Clearinghouse Document CD-16602.
Cameron, G. & Vanderwoerd, J. (1997).
Protecting children and supporting families. New York: Walter
De Gruyter, Inc.
Carrol, N.A. & Reich, J.W. (1978). Issues
in the implementation of the parent aide concept. Family Service
Association of America.
Dubowitz, H., Black, M., Kerr, M., Starr,
R., & Harrington, D. (2000). Fathers and child neglect,
Archives of Pediatrics and Adolescent Medicine, 154, 135-141.
Family Support America (n.d.). Retrieved
July 14, 2003, from http://www.familysupportamerica.org/content/learning_dir/principles.htm
Guterman, N. B. (2003). Tapping Parents
Informal Social Networks to Strengthen the Preventive Impact
of Home Visitation, presented at the Department of Defense
Joint Services New Parent Support Conference, New Orleans,
July.
Hanson, D. J., Pallotta, G. M., Tishelman,
A. C., Conaway, L., P., & MacMillan, V. M. (1989). Parental
problem-solving skills and child behavior problems: A comparison
of physically abusive, neglectful, clinic, and community families,
Journal of Family Violence, 4, 353-368.
Helfer, R.E. & Kempe, C.H. (1992). Helping
the battered child and his family. Philadelphia, PA: J.B.
Lippincott Company.
Hornick, J. & Clarke, M.E. (1986). A
cost/effectiveness evaluation of lay therapy treatment for
child abusing and high-risk parents. Child Abuse and Neglect
Journal, Volume 10.
King, G., Honaker, J., Joseph A. & Scheve,
K. (1999), Analyzing incomplete political science data: An
alternative algorithm for multiple imputation, Cambridge,
MA: Harvard University Department of Government.
Kitzman, H., Olds, D. L., Henderson, C.
R., Hanks, C., Cole, R., Tatelbaum, R., Mcconnochie, K. M.,
Sidora, K., Luckey, D. W., Shaver, D., Engelhart, K., James,
D., & Barnard, K. (1997). Effect of prenatal and infancy
home visitation by nurses on pregnancy outcomes, childhood
injuries, and repeated childbearing. Journal of the American
Medical Association, 278(8), 644-652.
Krell, H.L., Richardson, C., LaManna, T.N.
& Kairys, S.W. (1982). Parent aides as providers of secondary
preventive psychiatry. Volume 1, Number 4.
Leventhal, J. (1997). The prevention of
child abuse and neglect: Pipe dreams or possibilities? Clinical
Child Psychology and Psychiatry, 2(4), 489-500.
McKay, M. M., Stoewe, J., McCadam, K., &
Gonzales, J. J. (1998). Increasing access to child mental
health services for urban children and their caregivers, Health
and Social Work, 23(1), 9-15.
NPAN Brochures (2000). National parent aide
network –Bringing together programs to prevent abuse
and neglect. Toledo, OH: NECF.
Olds, D. L., Henderson, C. R., Chamberlin,
R., & Tatelbaum, R. (1986). Preventing child abuse and
neglect: A randomized trial of nurse home visitation. Pediatrics,
78, 65-78.
Pagani, S. & Quinones, I. (1996). ESCAPE
model for rural service delivery in San German, Puerto Rico,
evaluation report. CD-25095 National Clearinghouse on Abuse
and Neglect Information.
The David and Lucille Packard Foundation
(1999). The future of children, home visiting: recent program
evaluations. Volume 9, Number 1.
Polansky, N. A., Chalmers, M.A. , Buttenwieser,
E., & Williams, D. P.. (1978). Assessing adequacy of child
caring: An urban scale, Child Welfare, 57, 439-449.
Prinz, R. J., Smith, e. P., Dumas, J. E.,
Laughlin, J. E., White, D. W., & Barron, R. (2001). Recruitment
and retention of participants in prevention trials involving
family-based interventions, American Journal of Preventive
Medicine, 20, 1S, 31-37.
Rose, S. J. & Meezan, W. (1993). Defining
child neglect: Evolution, influences and issues, Social Service
Review, June, 279-293.
Straus, M. A., Hamby, S. L., Finkelhor,
D., Moore, D. W., & Runyan, D. (1998). Identification
of child maltreatment with the parent-child conflict tactics
scales: Development and psychometric data for a national sample
of American parents, Child Abuse and Neglect, 22, 249-270.
Turner, C. F., Ku, L. Rogers, S. M., Lindberg,
L. D., Pleck, J. H., & Sonnenstein, F. L. (1998). Adolescent
sexual behavior, drug use and violence: Increased reporting
with computer survey technology, Science, 280, 867-873.
U.S. Department of Health and Human Services.
(2003). Child Maltreatment 2001. Washington, D.C.: U.S. Government
Printing Office)
Vadasy, P., Project Director. (1991). Parent
aide & respite care programs: A profile of models for
prevention and family support. Seattle,
WA: Washington Research Institutes.
Vaux, A., Phillips, J., Holly, L., Tomson,
B., Williams, D., Stewart, D. (1986). The social support appraisals
scale: studies of reliability and validity, American Journal
of Community Psychology, 14, 195-219.
Wasik, B., Bryant, D. & Lyons, C. (1990).
Home visiting: Procedures for helping families. Newbury Park,
CA: SAGE Publications.
Yates, B. T. (1996). Analyzing Costs, Procedures,
Processes, and Outcomes in Human Services, Thousand Oaks:
Sage.
Zuravin, S. J. (1999). Child neglect: A
review of definitions and measurement research, in H. Dubowitz,
(ed.), Neglected Children: Research, Practice and Policy,
Thousand Oaks, CA: Sage, 24-46.
17.
Program Service Log
As part of the collection of data a log
detailing the types of service provided in person or on the
phone with the family was developed. The categories on the
log represent similar categorizations used in other research
on home visitors, ideas from our Parent Aide Staff and contributions
by other Exchange Center Parent Aide Directors who responded
from a post on the National Exchange Foundation List Serve.
Parent Aide staff first piloted this then modifications were
made. The log will be completed on every home visit or contact.
Dictation of the contact will be kept on the reverse of the
form for convenience (prior to the development of this form
parent aides just kept dictation).
While relationship is a key ingredient of
parent aide intervention there are many services that are
provided to address ecological concerns. It is hoped that
by detailing the number and types of services that were delivered
that we can correlate the effects of parent aide services
with more specificity. Our pre-service training, monthly in-service
trainings and weekly supervision equip the parent aides with
techniques and information in order to deliver the services.
Training on completion of this form includes
instructions in our Evaluation Manual and practice watching
videos of home interaction, rating and discussing the rating.
Through this process we hope to establish a high quality of
inter-rater reliability.
George Bryan
Project Director
-----------------------------------------------------------------------
[1] Cameron and Vanderwoerd, p. 147.
[2] Cameron and Vanderwoerd, p. 192.
|
